306
J. Agric. Food Chem. 1991, 39,306-310
Tissue Residues and Metabolism of Avilamycin in Swine and Rats John D. M a g n u s s e n , John E. Dalidowicz, Thomas D. Thomson, and Alvin L. Donoho' Animal Science Chemical Research, Lilly Research Laboratories, Division of Eli Lilly and Company, Greenfield, Indiana 46140
[14C]Avilamycinwas fed to growing swine at a level of 60-80 ppm (1.5-2 times the recommended use level), and tissues were assayed for radioactivity (RA). A t a practical zero withdrawal swine fed 60 ppm of uniformly labeled (UJ4C) avilamycin for 14 d a y s had RA residues of 0.14, 0.66,0.34, and 0.55 ppm in muscle, liver, kidney, and fat, respectively. Swine f e d 80 ppm of [14C]avilamycin labeled in the dichloroisoeverninic acid portion had residues 3-5 times lower, indieative that most of the residue was derived from the oligosaccharide portion of avilamycin. The primary metabolite in liver and feces was flambic acid. Most of the RA i n f a t from swine fed [UJ4C]avilamycin was i n the f a t t y acids. [I4C]Avilamycin was excreted rapidly and nearly quantitatively b y swine, with 5% of the dose i n the urine and the remainder i n feces. The excretion pattern and metabolic profile of [14C]avilamycin in the rat were similar to swine.
INTRODUCTION Avilamycin is an antibiotic of the orthosomycin family that consists of a six-member oligosaccharide, dichloroisoeverninic acid (DCIEA), and m e t h y l eurekanate. The avilamycin complex is produced b y Streptomyces uiridochromogenes and consists of factor A as the major component with several other minor components. T h e structures have been published (Mertz et al., 1986) and the p r i m a r y components are shown i n Figure 1. Avilamycin, when fed to swine, causes an increase in rate of gain and efficiency of feed utilization (Jones et al., 1987a,b;
Hw--cH3 xo 0
W a t k i n s e t al., 1987). MATERIALS AND METHODS Labeled Compound. [ l4CjAvilamycin was prepared by fermentation using two different substrates. [14C]Glucosesubstrate produced relatively uniformly labeled avilamycin. Hydrolysis and characterization of this material demonstrated that seven of the eight rings (all except ring D, Figure 1) contained radioactivity. Ring D is presumed to be labeled also, but the characterization was terminated short of confirmation of that labeling. A second substrate, [14C]diethylmalonate,produced avilamycin in which approximately 85% of the label was associated with the DCIEA ring and the remainder was in the rest of the molecule. The twolabeled materialswerecalled [U-WIavilamycin and [DCIEA-14C]avilamycin, respectively. [I4C]Avilamycin dosing materials were purified to 98% or greater radiochemical purity before use. Animal Feeding, Dosing, a n d Sample Collection. Balanceexcretion experiment SW-1was conducted to determine the rate, route, and extent of excretion of [14C]avilamycin. Two female pigs weighing approximately 40 kg were housed individually in metabolism cages and fed a standard grower ration containing 60 ppm of unlabeled avilamycin. On the day of dosing, the pigs were given a 120-mg dose of [U-14C]avilamycin,0.25 pCi/mg, mixed into 450 g of ration. After consumption of the dose, they were fed 450 g of unmedicated ration and were maintained on unmedicated ration for the remainder of the experiment. Urine and feces were collected quantitatively daily and frozen. For the tissue residue studies, crossbred barrows (castrate males) and females weighing 40-50 kg were housed individually in metabolism cages and conditioned on an unmedicated growerfinisher ration. The pigs were then fed twice daily a [14C]avilamycin-treated ration equivalent to nominal treatment levels of 80 or 60 ppm. The quantity of [14C]avilamycin needed daily was calculated on the basis of feed consumption of 4T of body
* A u t h o r to w h o m correspondence should be addressed. 002 1-8561191/ 1439-0306$02.50/0
FaCtOl A A' B E
F
G H I J
Other w
2 1 -CO-CH(CH3)2 -CO-CH2-CH3 - -COCH3 -H -CO-CH(CH3)2 -CO-CAHU -CO-CH(CH3)2 -CO-CHz-CH3 -CO-CH(CH3)2
,.
o
,
n
s
-HreDlaces .COCHa on Rino H CHOH-CH3 replaces -COCH3 on Ring H -OH replaces . O W 3 and H replaces the CI marked by * on Ring A -H replaces the Ci marked by * on Ring A -OH replaces -OCH3 on Ring F
F i g u r e 1. Structure of avilamycins. weight. This dose was administered ina portion of feed medicated a t twice the calculated treatment level. The pigs were allowed to consume the treated ration and then were given a portion of unmedicated ration. This regimen was used to ensure that the confined pigs actually consumed the appropriate dose. Treated feed was prepared by dissolving the appropriate quantity of [14C]avilamycin in 50-60 mL of acetone and pipetting this solution dropwise over the surface of 4 kg of control ration. The acetone was allowed to evaporate, and this "premix" was thoroughly mixed into the remainder of the ration. Prior to tissue sampling, the swine were sacrificed with a captive bolt device, exsanguinated as completely as possible, and then washed with a high-pressure hose to remove feed and fecal contamination. Samples of lumbar backfat and muscle (Longissimus dorsi) along with the liver and kidneys were then collected. Experiment SW-2 was conducted to determine the steadystate concentrations of RA in tissues of swine fed 80 ppm of [DCIEA-14C]avilamycin,0.242 pCi/mg. Nine pigs, five males @ 1991 American Chemical Society
J. Agric. Food Chem., Vol. 39,No. 2, 1991 307
Tissue ResMues and Metabolism of Avilamycin
and four females, were fed the treated ration. After dosing for 4 7 , or 10days, groups of three pigs were euthanized at a practical zero withdrawal time of 6 h after the last dose and tissues were taken for assay. Experiment SW-3 was conducted to determine the rate of decline of residue after withdrawal from treatment. Five pigs were fed the equivalent of 80 ppm of [DCIEA-14C]avilamycin, 0.249 pCi/mg, for 7 days and euthanized a t zero withdrawal (6 h), 3 days, or 5 days after the last dose. Experiment SW-4 was conducted to determine steady-state residues in tissues of swine fed 60 ppm of [U-14C]avilamycin.Six pigs were fed [U-14C]avilamycin,0.269 pCi/mg, and groups of three pigs were euthanized at zero withdrawal after dosing for 10 and 14 days. Rats were dosed with [Wlavilamycin to provide samples for characterization of radioactivity in excreta and tissues. In experiment RT-1, three male and three female Sprague-Dawley rats were housed in metabolism cages and dosed once daily for 3 days with [DCIEA-l4C]avilamycin,100 mg/kg of body weight, in acacia suspension. Urine and feces were collected quantitatively, through the 24-h period following the third dose. In experiment RT-2, three male and three female Sprague-Dawley rats were fed a ration containing 550 ppm of [U-14C]avilamycin for 4.5 days. Urine and feces were collected during dosing, and livers were taken from the rats a t zero withdrawal. S a m p l e Preparation a n d Analysis. Urine and feces were collected daily in the balance-excretion experiments. Feces samples were diluted with an equal weight of water and were homogenized to give a uniform paste. Tissues from residue experiments were ground and frozen until assayed. Five nominal 0.5-g replicates of muscle, liver, and kidney were each digested in 3 mL of NCS solubilizer (Amersham Corp.). After several days, a toluene-based scintillation cocktail was added for liquid scintillation counting (LSC). Five nominal l-g replicates of fat were melted a t approximately 80 “C in a counting vial and then combined with Aquasol (New England Nuclear Corp.) for LSC. Aliquots of urines were combined with a dioxane-Methyl Cellosolve-toluene based cocktail and assayed by LSC. Feces, extracted tissue solids, column packing materials, and feeds were assayed by combustion to 14C02,followed by LSC, as described by Herberg et al. (1978). Samples were counted in a Tri-Carb liquid scintillation counter (Packard Instruments Co.), and count rates were adjusted to 100°O efficiency by internal standardization with [14C]toluene. Tissues from untreated pigs were assayed to provide background data. The limit of detection (LOD) for tissue samples was defined as control counts per minute (CPM) + 3 standard deviations, or + 4 cpm, whichever was greater. In most experiments, recovery of RA was determined by assay of “spiked” control tissues to establish a limit of quantitation (LOQ). Characterization of Residues in Tissues a n d Excreta. Feces samples of 20 (swine) or 10 g (rats) were extracted by blending for 15 min in 200 mL of acetone. Each sample was filtered by vacuum filtration, and the filter cake was extracted under reflux for 30 min in 200 mL of 1:l acetone/water. The sample was filtered and fractions were assayed for RA. Liver samples of 50 (swine) or 5 g (rat) were extracted sequentially by blending in acetone and methanol and then by reflux in 1:l acetone/water. Solvent volumes were 4 and 20 mL/g for swine and rat samples, respectively. Column chromatography was performed with an 8 mm i.d. X 120 cm glass column packed with 50 mL of Woelm dry column silica gel (Universal Scientific, Inc., Atlanta, GA). The sample was dried onto 10 mL of the silica gel and packed into the column. A nonlinear gradient was developed in a 190-mL stirred reservoir filled initially with toluene. The elution was performed with 200 mL of each of the following solvents in sequence: 2:l toluene/ ethyl acetate (EA), 100RJ EA, 8:2 EA/methanol, 8:2:0.1 EA/ methanol/acetic acid, 6:4 EA/methanol, 100% methanol. Twenty-milliliter fractions were collected, and aliquots were assayed by LSC. Thin-layer chromatography (TLC) was performed on 20 X 20 cm silica gel 60 F ~ M plates (E. Merck) developed in solvent I (85:15:4 EA/cyclohexane/methanol), solvent I1 (95:5:2 EA/ methanol/acetic acid), or solvent I11 (9:l EA/methanol). Ra-
Table I. Excretion of Single Doses of [W]Avilamycin by Swine % of administered dose pig 594 pig 596 days postdose urine feces urine feces mean cumulative 1 2.8 0.1 3.3 0.0 3.1 2 0.7 65.7 0.9 18.2 45.9 94.8 0.3 72.8 3 0.2 24.5 4 0.3 1.6 0.0 2.2 96.8 5 0.1 0.5 0.3 0.5 97.5 98.0 0.0 0.4 0.2 0.4 6-9 total total excreted
4.1
92.8
96.9
5.0
94.1
99.1
98.0
dioactive zones were visualized on TLC plates by autoradiography with X-OMat film (Kodak). Selected samples were assayed for DCIEA by basic hydrolysis of extracts and gas-liquid chromatography (GLC) assay of the DCIEA after methylation with diazomethane. GLC was performed with a Hewlett-Packard Model 5730 instrument equipped with electron capture detector. The column was 37; Carbowax 20M on Chromosorb W-HP, eluted with 10% methane in argon at an oven temperature of 200 “C. Fat samples were melted in beakers on a hot plate, decanted from connective tissue, cooled, and assayed by LSC. Aliquots were saponified at 55 “C in 5 mL/g 50% aqueous NaOH for 4-6 h, followed by dilution with 2 volumes of water and heating for an additional 40 h. A sample equivalent to 2 g of melted fat was washed into a separator with 120 mL of water and extracted with 2 X 100 mL of CHC13. Fatty acids were extracted into 100 + 50 + 50 mL of CHClB after adjustment to pH 3. Nominal 0.5-g samples of fatty acids were dissolved in 50 mL of absolute ethanol, neutralized with 2 N NaOH, and evaporated to ca. 30 mL to remove traces of water. The p-bromophenacyl esters were prepared by reaction with 0.5 g of p-bromophenacyl bromide under reflux for 2 h. The reaction mixture was cooled and diluted with 100 mL of hexane, 180 mL of methanol, and 40 mL of water. The hexane was removed after separation of phases, and the aqueous alcohol was extracted additionally with 100 + 50 + 50 mL of hexane. The combined hexanes containing the phenacyl esters were purified by recrystallization from methanol. The progress of purification was monitored by HPLC using a Waters FBondapak CIS 4.6 mm X 25 cm column eluted with 9 5 5 methanol/water with detection a t 254 nm. RESULTS
Excretion of [14C]Avilamycin. The rate, route, and extent of excretion of [14C]avilamycin b y swine i n experi m e n t SW-1 are shown i n T a b l e I. Recovery of the dose averaged 98% for the two pigs. Approximately 93% of the dose was excreted i n the feces, and 5 % was excreted i n the urine. Approximately 95% of the administered dose was excreted within 3 days after dosing. In rat metabolism experiment RT-1, urine contained less than 0.596 of the administered dose, while feces contained 96 % (n = 6 animals). Tissue Radioactivity Concentrations. Experiment S W-2. Results f r o m the [DCIEA-14C]avilamycin steadystate study, expressed as avilamycin equivalents i n tissues, a r e s h o w n in T a b l e 11. There was n o detectable residue of RA in muscle (C0.02 ppm). Liver, kidney, and fat contained 0.2, 0.1, and 0.1 p p m , respectively. RA concentrations achieved steady state i n liver and kidney after 7 d a y s of dosing since 10-day concentrations were not significantly higher (one way ANOVA, P > 0.05) t h a n 7-day concentrations. There was a suggestion of progressively higher RA i n f a t with longer dosing times. Liver, kidney, and f a t f r o m the three pigs i n the 10-day dose group were assayed for parent avilamycin by thin-layer bioautograp h y and for DCIEA b y GLC. K i d n e y a n d f a t contained
308
J. Agric. Food Chem., Vol. 39, No. 2, 1991
Magnussen et ai.
Table 11. Residues of Radioactivity in Tissues of Pigs Fed 80 ppm of [DCIEA-l4CC]Avilamycinfor 4-10 Days and Euthanized at Zero Withdrawal
animals dosing interval, days muscle 2 males, 1 female 4 NDR" 1 male, 2 females 7 NDR 2 males, 1 female 10 NDR a No detectable residue at a limit of detection of 0.02 ppm. Table 111. Residues of Radioactivity in Tissues of Pigs Fed 80 ppm of [DCIEA-14C]Avilamycinfor 7 Days and Euthanized at Various Withdrawals
animals female male female male female a
withdrawal, days 0
3 5
ppm net radioactivity muscle liver kidney fat NDRO 0.147 0.077 0.067 NDR NDR 0.026 0.044 NDR NDR 0.022 0.062 NDR NDR NDR 0.047 NDR NDR 0.025 0.049
No detectable residue at a limit of detection of 0.02 ppm.
no detectable residue of parent avilamycin (LOQ of 0.05 ppm). Liver contained traces of avilamycin, but concentrations were below 0.05 ppm. Kidney and fat contained no detectable DCIEA (LOQ of 0.1 ppm), while liver contained a mean of 0.126 ppm. Thus, most of the liver RA consisted of residue containing the DCIEA moiety, but very little was parent avilamycin. Experiment SW-3. Results from the [DCIEA-14C]avilamycin tissue residue decline study are shown in Table 111. The residue pattern at zero withdrawal was consistent with the previous experiment. Residues in liver and kidney declined to near nondetectable concentrations by 3 days. Residue in fat declined more slowly. Experiment S W-4. Results from the [U-14C]avilamycin steady-state study are shown in Table IV. Muscle, liver, and kidney residues were approximately 0.1,0.6, and 0.3 ppm, respectively. Residues after 14 days of dosing were not significantly higher than for 10 days. The residues in fat were statistically higher in the 14-day group (0.55 vs 0.26 ppm). Comparison of these data with experiment SW-2 shows that dosing with uniformly labeled avilamycin gave residues 3-5 times higher than when the label was primarily in the DCIEA portion of the molecule. Characterization of RA i n Feces a n d Urine. Twenty-gram samples of feces from swine fed [DCIEA14C]avilamycin (experiment SW-2) and swine fed [UJ4C]avilamycin (experiment SW-4) and 10 g of feces from rats fed [U-14C]avilamycin(experiment RT-1) were extracted as described under Materials and Methods. Silica gel column chromatograms were generated for each feces sample on pooled 10% aliquots of the acetone and acetone/ water extracts. A comparison of the distribution of RA is shown in Figure 2, and the chromatograms are compared in Figure 3. More than 95 % of the RA was extracted from each sample, and the extractable RA gave qualitatively similar radiochromatograms with three major peaks (Figure 3). The distribution of RA and chromatographic profile was quantitatively similar for the [U-14C]avilamycintreated pigs and rats. The feces from the [DCIEA-14C]avilamycin-treated pigs yielded most of the residue in the acetonelwater extract. The chromatogram from this sample indicated that most of the sample RA was in peak 3. Pools of fractions from peaks 1-3 were prepared for each feces sample, and these were evaluated by TLC. Results are shown in Figure 4. Peak 1is the region where parent avilamycin elutes from the silica gel column. TLC
ppm net radioactivity (mean f SD) liver kidney 0.211 f 0.029 0.226 f 0.035 0.216 f 0.028
0.099 f 0.005 0.097 f 0.005 0.102 f 0.006
fat 0.057 f 0.012 0.080 f 0.007 0.121 f 0.002
of peak 1 samples demonstrated that the peak consisted primarily of two components, avilamycin factor A at an Rf of ca. 0.6 and an unidentified compound at Rf of ca. 0.4. There was good qualitative agreement among all three feces samples. Avilamycin constituted approximately 8 % of the fecal RA from the U-14C-treated pig. The unidentified compound had an apparent MW of 1404, 2 mass units greater than that of avilamycin, by negative fast atom bombardment mass spectrometry. It was not characterized further since it was not a major metabolite in tissues of treated swine. TLC of peak 2 samples demonstrated a qualitative difference between U-14C-and DCIEA-14C-treated animals. Feces fractions from swine and rats treated with [U-14C]avilamycin contained three metabolites which moved just off the point of application (origin). In contrast, the RA in the sample from the DCIEA-14C-treated swine was primarily polar material which remained at the origin. These results indicate that the three metabolites were derived from the oligosaccharide and/or eurekanate portion of avilamycin. TLC of peak 3 samples demonstrated that most of the RA was in a single compound. It was relatively polar and required an acidic TLC solvent for mobility. This compound was subsequently determined to be flambic acid (Figure 5 ) . It was relatively unstable and cyclized readily to flambalactone, which has a mobility slightly greater than avilamycin. Quantitation by counting silica gel segments scraped from the plate indicated that these two zones constituted 80% or more of the peak 3 RA for both swine and rats. Similar column and TLC fractionations were conducted on urines from [U-14C]avilamycin-treatedswine and rats. The results indicated the presence of the same primary metabolites as were observed in feces samples. Characterization of RA in Livers a n d Kidneys of Swine a n d Rats. Preliminary characterization was done on liver from [DCIEA-14C]avilamycin treated pigs to determine the residue profile. This work indicated that there was very little parent avilamycin present and that the only major metabolite was a nonpolar compound which was identified as flambalactone. Two minor metabolites were observed by TLC. One had a TLC mobility similar to that of the unidentified peak 1 metabolite from feces, and the other was a polar acidic compound. Further work indicated that the polar acidic compound was flambic acid (Figure 5) and that it was undergoing conversion to flambalactone during sample processing. Subsequently, fractionation was performed on livers from [ U-l4C]avi1amycin-treated pigs and rats by procedures described under Materials and Methods. Liver samples were extracted with acetone, methanol, and 1:l acetonelwater. The distribution of radioactivity from six individual pigs and a pool of rat livers is shown in Table V. Representative column radiochromatograms from one male and one female pig and the female rat pool are shown in Figure 6. The liver chromatograms had three distinct features. There were the peak 1, peak 2, and peak 3 regions
309
J. Agric. Food Chem., Vol. 39, No. 2, 1991
Tissue ResMues and Metabolism of AvilamychI
Table IV. Residues of Radioactivity in Tissues of Pigs Fed 60 ppm of [U-W]Avilamycin ppm net radioactivity (mean f SD) animals 2 males, 1 female 2 males, 1 female
dosing interval, days 10 14
muscle 0.093 f 0.036 0.135 f 0.031
Feces
liver 0.554 f 0.143 0.659 f 0.159
kidney 0.321 f 0.083 0.340 f 0.010 Peak 2
Peak 1
Peak 3 Solvent System It Std. Pig 1 Pig 2 Rat
Solvent System 111
Solvent System I a d . Pig 1 Pig 2 Rat
Acetone
fat 0.261 f 0.094 0.551 f 0.034
Pig 2 Rat
1
Acetone Extract Sample 1 25% Sample 2 71% Sample 3 78%
3
Filter Cake Acctone:Water Ref lux Sample ExtL 1 71% Sample 2 27% Sample 3 22%
Spent SampleFeces 1 4% Sample 2 2% Sample 3 ~ 1 %
10% Aliquot to Silica gel Column
. .
Figure 2. Distributionof radioactivityin feces extracts. Samples 1-3 are [DCIEA-14C]avilamycin-treated pig, [U-14C]avilamycintreated pig, and [ U-14C]avilamycin-treatedrat, respectively.
1 1 . .
,
Figure 4. Autoradiograms from silica gel column peaks 1-3. Pig 1, [ DCIEA-14C]avilamycin-fed pig; pig 2, [U-14C]avilamycin-fed Pig.
H9. P
‘--OH
CH3
b
A
CH3O
p
‘0IO
H
cH3$./.
CI OH
Figure 5. Structures of flambic acid and its derivatives. Flambic acid, R = H; methyl flambate, R = CHs; flambalactone, ROH removed to form lactone on ring C. 0
10
20
30 40 Fraction Number
50
60
Figure3. Silica gel column radiochromatograms of feces extracts.
qualitatively similar to the feces samples. There was a nonpolar region in fractions 1-10 and a polar region after peak 3. In both of the swine liver samples the peak 1RA was less than 0.02 ppm, which demonstrates that very little of the RA was parent avilamycin. This result is consistent with the results from experiment SW-2 in which only traces
